Add monadic variants for iterateN, unfoldr, unfoldrN

[Shimuuar]

I needed such functions while working on some toy Monte-Carlo simulations. It turned out that all required stream fusion machniery for these functions was present already but they weren't implemented for some reason

There're no tests. I don't really know what to test here.

[cartazio]

a starting point for writing tests would be some tiny examples and checking that the results are as expected

[Shimuuar]

Only property I can think of is that iterateM produces same results as iterate for identity monad. But that should be true by construction. It probably won't hurt t add such test but it wouldn't be of much help

[cartazio]

toy inputs plus results to make sure they're in the test suite will still be nice. just bog simple "heres an input and the expected result" level testing is better than no testing at all! ;)

[Shimuuar]

I've added tests for new functions and restored some commened out tests. Test suite does need some love.

P.S. I think this shoudl fix #8

[cartazio]

great, THANK YOU.

(NB: i havent done a close reading of those tests to make sure they're good, I'm just expressing gratitude rather than indicating i've done any form of code review)

[Daniel-Diaz]

Bump. I'm interested in this too!

[cartazio]

this is a nice PR, we're catching up on stuff, but will look at it in the next week or so. if you find some time to clean it it up for master please let me know, else happy to have a go myself :)

[Shimuuar]

Will do that

[cartazio]

yay :)
i may want some help de bit rotting the benchmark suite once @dolio finishes moving to the new style CI setup, at which point we could add this stuff to the benchmark suite.

@Shimuuar hope you dont mind that side diversion :0

[Shimuuar]

I've rebased branch to the current master.

[cartazio]

Thanks! We'll review it soon! Awesome

On Monday, July 25, 2016, Aleksey Khudyakov notifications@github.com
wrote:

I've rebased branch to the current master.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#41 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAAQwj_GGxulSr6n1BMsE5mo9-DWhkS3ks5qZQR7gaJpZM4Cimwp
.

[fhaust]

As this was apparently merged into master, has this ever been released?

[glguy]

No, you you look at http://hackage.haskell.org/package/vector you'll see the upload date is earlier than the merge.